Narcotic therapy forms the mainstay of pain management. These drugs can be administered in many forms to patients with postsurgical and other forms of acute and chronic pain. Morphine, one of the oldest narcotics, is available for administration in tablet or in injectable form. Fentanyl, a synthetic narcotic, was first synthesized in 1960 by Paul Janssen and found to be 150 times more potent than morphine [Theodore Stanley, "The History and Development of the Fentanyl Series," Journal of Pain and Symptom Management (1992) 7:3 (suppl.), S3-S7]. Fentanyl and its relatives Sufentanil and Alfentanil are available for delivery by injection. In addition, fentanyl is available for administration by a transdermal delivery system in the form of a skin patch [Duragesic.TM. (fentanyl transdermal system) package insert, Janssen Pharmaceutica, Piscataway, N.J. 08855, January-June 1991].
A feature of the synthetic narcotic fentanyl is that is has a more rapid time to onset and a shorter duration of action than morphine. This makes fentanyl a useful drug for the management of acute pain. Currently, fentanyl is typically given by intravenous injection for acute pain management. Although fentanyl can be given by a transdermal patch, transdermal delivery of fentanyl is designed for long-term administration of the drug and does not lend itself to achieving a peak level rapidly for a short-term effect.
An alternative to delivery by injection for narcotics is delivery by inhalation. Morphine [J. Chrusbasik et al., "Absorption and Bioavailability of Nebulized Morphine," Br. J. Anaesth. (1988) 61, 228-30], fentanyl [M. H. Worsley et al., "Inhaled Fentanyl as a Method of Analgesia," Anaesthesia (1990) 45, 449-51], and sufentanil [A. B. Jaffe et al., "Rats Self-administer Sufentanil in Aerosol Form," Psychopharmacology, (1989) 99, 289-93] have been shown to be deliverable as aerosols into the lung. The pilot study described by Worsley suggested that "inhaled fentanyl is an effective, safe and convenient method of analgesia which merits further investigation."
Inhalation of a potent synthetic narcotic aerosol provides a mechanism for the non-invasive delivery of rapid-acting boluses of narcotic. The on-demand administration of boluses of narcotic coupled with a controlled baseline intravenous infusion of narcotic is termed "patient-controlled analgesia" (PCA) and has been found to be a very effective means of postoperative pain management.
Demand analgesia was first introduced in 1968 by Schetzer who showed it to be an effective mechanism for treating postoperative patients [Maureen Smythe, "Patient-Controlled Analgesia: A Review," Pharmacotherapy (1992), 12:2, 132-43]. Prior to the availability of patient-controlled analgesia, the paradigm for postoperative pain management consisted of intermittent intramuscular injections of narcotic. The cycle of the patient felling pain, calling the nurse who then must locate and bring the drug to the bedside for administration results in suboptimal postoperative pain management [Philip Shade, "Patient-controlled Analgesia: Can Client Education Improve Outcomes?," Journal of Advanced Nursing (1992) 17, 408-13]. Postoperative pain management by intermittent narcotic administration has been shown to be a largely ineffective method of pain management for many of the patients undergoing the more than 21 million surgical procedures in the United States each year [John Camp, "Patient-Controlled Analgesia," AFP (1991), 2145-2150]. Even if every patient reliably received a constant dose of narcotic postoperatively, studies of therapeutic narcotic pharmacokinetic data have shown that patient variability makes such an approach fundamentally unsound and potentially dangerous [L. E. Mather, "Pharmacokinetics and Patient-Controlled Analgesia," Acta Anaesthesiologica Belgica (1992) 43:1, 5-20].
The first commercial device for automatically providing intravenous patient-controlled analgesia was developed in Wales in the mid-1970s. This device, the Cardiff Palliator (Graesby Medical Limited, United Kingdom) is the predecessor of numerous currently available computer-controlled patient-controlled analgesia intravenous pumps [Elizabeth Ryder, "All about Patient-Controlled Analgesia," Journal of Intravenous Nursing (1991) 14, 372-81]. Studies using these computer controlled intravenous narcotic infusion pumps have shown that small doses of narcotics given on demand by the patient provided superior pain relief when compared with intermittent intramuscular administration of these drugs [Morton Rosenburg, "Patient-Controlled Analgesia," J. Oral Maxillofac Surg (1992) 50, 386-89].
These computer-controlled pumps typically allowed for the programming of four different parameters: 1) basal intravenous narcotic infusion rate; 2) the bolus of narcotic to be delivered on each patient demand; 3) the maximum hourly total dose of narcotic to be allowed; and 4) the lockout period between doses. Typical programming for postoperative pain management with intravenous fentanyl might be a basal infusion rate of 20 .mu.g/hr, a bolus demand dose of 20 .mu.g, a maximum hourly dose of 180 .mu.g, and a lockout period between doses of 5 minutes. In a study of 30 patients treated for postoperative pain with intravenous fentanyl patient-controlled analgesia, the minimum effective concentration (MEC) of fentanyl in the blood required to achieve pain relief in the group of patients studies was found to range from 0.23 to 1.18 ng/ml. Clinically significant respiratory depression was not seen in this study consistent with published data indicating that a fentanyl concentration of 2 ng/ml in the blood is typically required to depress the respiratory rate [Geoffrey Gourlay et al., "Fentanyl Blood Concentration--Analgesic Response Relationship in the treatment of Postoperative Pain," Anesth Analg (1988) 67, 329-37].
The administration of narcotic for pain management is potentially dangerous because overdoses of narcotics will cause complications such as respiratory depression. The patient's respiratory rate is decreased by the administration of narcotics. This decrease in respiratory rate may not be associated with a change in respiratory tidal volume [Miller, Anesthesia (2nd ed), Churchill Livingston, I, 762]. The four programmable parameters available on computer-controlled intravenous patient-controlled analgesia infusion pumps must be selected so as to minimize the likelihood of narcotic overdose. The preferred technique is to set the basal infusion rate at a relatively low rate and increase this rate based on how many times the patient presses the bolus demand button to self-administer supplemental drug.
As long as the patient himself or herself is the only one to push the demand button, respiratory depression is unlikely. However, there have been documented cases of the patient's family and friends pressing the narcotic demand button, for instance while the patient is sleeping [Robert Rapp et al., "Patient-controlled Analgesia: A Review of the Effectiveness of Therapy and an Evaluation of Currently Available Devices," DICP, The Annals of Pharmacotherapy (1989) 23, 899-9040].
It is a problem with patient-controlled analgesia that it must currently be performed using an intravenous infusion pump. This requires that an indwelling catheter be placed in the patient's vein and that the patient transport a relatively bulky system with himself at all times to receive a baseline infusion of intravenous narcotic and allow for intermittent on-demand self-bolusing of additional narcotic in order to match the patient's changing need for drug. A portable PCA device incorporating a wristwatch-like interface has been described [D. J. Rowbotham, "A Disposable Device for Patient-Controlled Analgesia with Fentanyl," Anaesthesia (1989) 44, 922-24]. This system incorporated some of the features of computer-controlled programmable PCA infusion pumps such as basal infusion rate and the amount of each bolus. However, this system, which involved the use of an intravenous catheter as seen in larger infusion pumps, incorporated no provision to record accurately the actual dose of Fentanyl administered to the patient over time.
Although fentanyl can be administered by transdermal patch, this method has been found to be suboptimal for postoperative main management [K. A. Lehmann et al., "Transdermal Fentanyl for the Treatment of Pain after Major Urological Operations, Eur. J. Clin Pharmacol (1991) 21:17-21]. Lehmann found that the low dose of narcotic delivered by transdermal fentanyl was inadequate to provide pain relief to many of his patients and that boosting the baseline infusion rate of the patch would put some patients at risk for having significant respiratory depression. In addition, he points out that if such a complication were to appear in conjunction with the delivery of narcotic by transdermal patch, the infusion could not be quickly stopped because the "cutaneous fentanyl depot" created by the transdermal patch would cause narcotic infusion to continue even after removal of the patch.
Delivery of fentanyl by aerosol used in conjunction with a non-invasively delivered long-acting preparation of narcotic such as slow-release oral morphine or a fentanyl transdermal patch provides a means for non-invasive administration of a basal rate of narcotic and rapid-acting boluses of narcotic to an ambulatory patient.
It is a problem with the aerosol delivery of fentanyl previously described that inefficient, bulky nebulizers must be used for the administration of the drug. In addition, these nebulizers work by administering from an open reservoir of the drug in aqueous solution allowing the vapor to be generally distributed and creating the potential for overdosing due to the lack of reproducible aerosol delivery. In addition, abuse through theft of the aqueous-phase fentanyl and subsequent unauthorized repackaging of this controlled substance in an aqueous injectable form are possible.
Because most surgery today is being done on ambulatory patients and because these patients are often rapidly discharged from the hospital and because patient-controlled analgesia has been identified as the preferred method of postoperative pain management, it is desirable to have a safe and effective method for non-invasive, ambulatory patient-controlled analgesia.